Therapy for malignant astrocytomas at UCSF has been surgery followed by adjuvant radiation therapy and chemotherapy. This proposal uses a genetic approach for malignant astrocytoma evaluation and treatment. We target radiation therapy for analysis because it is the most successful of the adjuvant therapies, and because there are three clear groups of patients: those whose tumors progress through their radiation therapy, patients whose tumors stabilize during radiation therapy, and those whose tumors decrease in size. We hypothesize that a tumor's radiation sensitivity depends upon genetic factors. Thus, the loss or gain of genetic material at any locus that determines radiation sensitivity should influence how a tumor responds to radiation. Because there are many possible sites for genetic regulation of radiation response, the first stage of the proposal encompasses several studies of primary astrocytic tumors and cell lines to determine a broad range of genetic pathology in astrocytic brain tumors. This knowledge of genetic alterations in different grades of astrocytic neoplasms should lead to an understanding of which lesions are associated with initiation and which are associated with progression. In the second stage, we will ask whether any of these lesions contribute to clinical radiation response in patients. The specific aims are: 1) to identify and map regions of loss or gain in DNA from glioma cell lines, primary human gliomas, and primary and recurrent malignant glioma pairs; 2) to compare radiation- sensitive and radiation-resistant tumors for losses and gains of genetic material at loci defined in Aim 1; and, 3) to correlate this genetic data obtained in Aims 1 and 2 with clinical and laboratory observations using the Core database.